The Phenomenon of “Fluid Creep” in Acute Burn … Phenomenon of “Fluid Creep” in Acute Burn Resuscitation ... The development of effective fluid resuscitation reg- ... Journal

ORIGINAL ARTICLES

The Phenomenon of “Fluid Creep” in AcuteBurn Resuscitation

Jeffrey R. Saffle, MD, FACS

Several reports have documented that modern burn patients receive far more resuscitationfluid than predicted by the Parkland formula—a phenomenon termed “fluid creep.” Thisarticle reviews the incidence, consequences, and possible etiologies of fluid creep in modernpractice and uses this information to propose some therapeutic strategies to reduce or elimi-nate excessive fluid resuscitation in burn care. A literature review was performed of histori-cal references that form the foundation of modern fluid resuscitation, as well as reports offluid creep and its consequences. The original Parkland formula required a 24-hour volumeof 4 ml/kg/%TBSA lactated Ringer’s solution followed by an infusion of 0.3–0.5 ml/kg/%TBSA plasma. Modern iterations of this formula have omitted the colloid bolus. Numer-ous exceptions to the formula have been noted, most consistently patients with inhalationinjuries. In contrast, recent reports document greatly increased fluid requirements in un-selected patients, which seems to consist largely of progressive edema formation in un-burned areas, increasing after the first 8 hours post-burn. This has been linked to occur-rence of the abdominal compartment syndrome and other serious complications. Strategiesto reduce fluid creep include the avoidance of early overresuscitation, use of colloid as aroutine component of resuscitation or for “rescue,” and adherence to protocols for fluidresuscitation. Fluid creep is a significant problem in modern burn care. Review of originalinvestigations of burn shock, coupled with modern reports of fluid creep, suggests severalmechanisms by which this problem can be controlled. Prospective trials of these therapiesare needed to confirm their effectiveness. (J Burn Care Res 2007;28:382–395)

The development of effective fluid resuscitation reg-imens is one of the cornerstones of modern burntreatment and perhaps the advance which has mostdirectly improved patient survival. At the beginningof World War II, patients with even moderate burnsoften died within a few days, of progressive shock andrenal failure. In 1921, Underhill’s study of victims ofthe Rialto Theater fire led him to conclude that loss ofintravascular volume led to a life-threatening “shock-like” syndrome that should be treated with infusionsof normal saline.1 In 1942, Cope and Moore de-signed the first formal resuscitation regimen to treatvictims of the Cocoanut Grove nightclub fire, withdemonstrated reduction in mortality.2–4 With contin-

ued refinements in resuscitation, almost all patients cannow be resuscitated successfully, and renal failure com-plicating acute burn injury has become rare.

A host of formulas have been used for burn resus-citation, almost all based on body weight and burnsize, and using various combinations of fluids. Thearchetype for such regimens and unquestionably themost widely used is the Parkland formula, describedby Charles Baxter.5 During the interval—now almost40 years—since publication of this formula, its accu-racy has become universally accepted in the burn carecommunity. It has been very surprising, therefore,that recent reviews have repeatedly demonstratedthat patients with major burns now often require re-suscitation volumes which significantly exceed Park-land predictions.6–11 The explanation for this experi-ence is unclear, but its occurrence has been linked toincreased recognition of the complications of edema,including the abdominal compartment syndrome(ACS), and been viewed with alarm by clinicians.Pruitt coined the term “fluid creep” to describe thisinsidious trend, and called for clinicians to “push the

From the Department of Surgery, University of Utah HealthCenter, Salt Lake City, Utah.

Address correspondence to Jeffrey R. Saffle, MD, FACS,Department of Surgery, 3B-306, University of Utah HealthCenter, 50 N. Medical Drive, Salt Lake City, UT 84132.

Copyright © 2007 by the American Burn Association.1559-047X/2007

DOI: 10.1097/BCR.0B013E318053D3A1

382

pendulum back” in the direction of more conserva-tive resuscitation.12

Is “fluid creep” really a new phenomenon, or hasthe intimidating stature of the Parkland formula keptclinicians from speaking up to challenge one of themost cherished icons of burn treatment until recently?Has the nature of burn injury changed, as Engrav et alhave suggested,11 or have clinicians been influenced byattitudes in other areas of medicine to practice fluidtherapy differently? Most important, is “fluid creep”harmful to patients and can it be prevented or con-trolled? This review addresses these questions.

EVOLUTION OF THEPARKLAND FORMULA

The period from 1965 to 1980 was a time of unprec-edented progress in burn treatment: effective topicalantibiotics were introduced, successful nutritional sup-port was pioneered, early excision was popularized,and resuscitation regimens were perfected.13 Manynow-legendary physicians contributed to our under-standing of the pathophysiology of burn shock, in-cluding Evans,14 Moncrief,15 Moyer,16 Arturson,17

Pruitt,18 Monafo,19 Shires, and others.In 1968, Baxter reported that resuscitation of

dogs with 50% TBSA burns with a volume of lactatedRinger’s solution (LR) equal to 24% to 32% bodyweight returned cardiac output and extracellular fluid(ECF) and plasma volumes to near normal, restoredtranscellular membrane potentials, and correctedmetabolic and lactic acidosis by the end of 24 hours.20

Optimal results were achieved when most of this fluidwas given in the first 8 hours after injury. A trial groupof 11 patients with burns of 30% to 85% TBSA weresimilarly resuscitated and required a volume of 3.5 to4.5 ml of LR per kilogram body weight, per percentTBSA burned (ml/kg/%TBSA) over the first 24hours. Baxter noted that crystalloid alone, however,would not completely replete ECF volume; some col-loid replacement was also needed to accomplish this.

These observations formed the basis of the originalParkland formula. It called for LR at a volume of 4.0ml/kg/%TBSA for the first 24 hours. Half was givenin the first 8 hours after injury and the rest was givenover the next 16 hours and adjusted to maintain urineoutput. Importantly, Baxter’s formula also included afourth 8-hour period during which plasma, at a vol-ume of 0.3 to 0.5 ml/kg/%TBSA, was given to com-plete resuscitation. During the remainder of the sec-ond 24-hour period, dextrose and water were given asneeded to maintain urine output.21

In 1979, Baxter reported results of this formula inthe resuscitation of 954 patients treated from 1973 to

1977.22 He found that 70% of 438 adults and 98% of516 children were resuscitated successfully, with 24-hour volumes ranging from 3.7 to 4.3 ml/kg/%TBSA. Only 12% of adults required more than thisvolume, whereas 18% required less. He emphasizedthe importance of restoring cardiac output with theuse of plasma, noting that output tended to level offat a low-normal level and that “further increases areunusual until plasma is administered in the 4th 8-hourperiod.” In experiments in which patients were givenboluses of plasma at various times post-injury, plasmawas found to be most effective in restoring ECF vol-ume if given after 24 hours post-burn.23

These principles and their results were reiteratedrepeatedly in the next few years and were combinedwith recommendations from other burn centers.The formula represented an improvement over theearlier Brooke15 and Evans14 formulas. As reviewedby Moncrief,24 all three formulas were designed tocontinue resuscitation though an initial 48-hourperiod; all resulted in administration of roughlyequivalent amounts of sodium; and all relied oncolloid administration as an important adjunct toreplete plasma volume.

Also in 1979, an NIH-sponsored conference onburn care was summarized with a statement that burnpatients should be resuscitated with as little fluid aspossible to maintain organ perfusion. Initial fluidtherapy should consist of isotonic crystalloid at a vol-ume between 2 and 4 ml/kg/%TBSA for the first 24hours and titrated to maintain urine output of 30 to50 ml/hr.25 The use of colloid in the second 24-hourperiod was not included. This recommendation hasstood as the accepted “consensus” for burn resusci-tation for over 25 years25,26 and has resulted in theconcurrent definition of the Parkland formula as amethod to predict fluid requirements in the first 24hours only, and without the use of supplementarycolloid.27,28 This departure from the original Park-land formula may help explain the occurrence offluid creep.

OBSERVATIONS OF FLUID CREEP

Even before the Parkland formula was published, ex-ceptions began to be identified. Baxter found that theformula would successfully resuscitate most burn pa-tients but noted that response was variable and thatadjustment in infusion rates on the basis of responsewas essential. He identified some patient groupswho routinely required additional fluid. These in-cluded patients with inhalation injuries, patientswith electrical burns, and those in whom resuscita-tion was delayed.5,29

Journal of Burn Care & ResearchVolume 28, Number 3 Saffle 383

Inhalation InjuriesIn a subsequent NIH consensus conference in 1981,both Baxter and Pruitt identified this group as re-quiring increased resuscitation.30,31 This observationhas been confirmed repeatedly, as reviewed in Table1.32–36 The absolute quantity of fluid required hasvaried substantially between studies, possibly de-pending on other variables, such as the use of colloid.For example, Navar et al used only crystalloid foracute resuscitation, and their patients required morefluid than did the patients reviewed by Herndon’sgroup, which has historically used a combination ofcolloid and LR for initial resuscitation.37 Hughes et alresuscitated their patients entirely with plasma pro-tein fraction but still found a roughly equivalent in-crease in relative fluid requirements.38 Thus, everyreview has confirmed that patients with inhalationinjury require an increase in fluid requirements, from35% to 65% above those of patients without inhala-tion injury, regardless of the regimen used.

Other Patient GroupsSubsequently, a number of other patients have beenidentified who are acknowledged to require fluidresuscitation that can significantly exceed Parklandpredictions. In addition to inhalation injury, the listincludes patients with secondary injuries, includingmultiple trauma and electrical burns; patients inwhom the onset of resuscitation is substantially de-layed39; and patients with alcohol or drug addic-tion.40,41 In addition, inexperienced clinicians oftenmake substantial errors in estimating burn extent anddepth, which can result in significant under- or over-calculation of fluid requirements.42 However, al-though these exceptions have been widely recog-nized, the accuracy of Parkland-based resuscitationfor the majority of patients has not been impugned bythese reports. Rather, these experiences have beenused to emphasize the necessity of monitoring pa-tients carefully and adjusting fluid infusions on thebasis of patients’ response.

Table 1. Comparison of fluid requirements in adult patients with and without inhalation injury

Reference

No. ofPatients,Group Resuscitation Required Difference Comment

Baxter (1981)30 NA, INH 5.37 l/m2 TBSA Review of fluid requirements in adult burninjuries.NA, Non-INH 3.31 l/m2 TBSA 62%

Scheulen andMunster(1982)34

48, INH 14.5 l/24 hours (10.6 predicted) �37% Review of adults with burns of 10% to 60%TBSA.53, Non-INH 9.7 l/24 hours (9.2 predicted) �5%

Navar et al(1985)33

51, INH120, Non-INH

5.76 � 0.39ml/kg/%burn3.98 � 0.39ml/kg/%burn

45% Retrospective review of children and adults withburns of �25% TBSA. Patients with INH tooklonger to resuscitate (29.8 � 1.3 vs 23.8 �

0.74 hours; P � .05). No colloids used.

Herndon et al(1988)32

20, INH14, Non-INH

3.8 � 1.5ml/kg/%burn2.3 � 1.2 ml/kg/%burn

65% All adults with major burns. Lactated Ringer’ssolution alone used during the first 24 hours.Colloids not mentioned.

Hughes et al(1989)38

9, INH 4.38 � 1.26ml/kg/%burn 29% Review from Britain. All patients resuscitatedentirely with plasma protein fraction.Requirements exceeded predicted value of3.3 ml/kg/%TBSA.

Darling et al(1996)36

100, INH 6.52 � 0.26ml/kg/%burn Fluids received increased the odds ratio fordeath by 1.18 and odds ratio for acuterespiratory distress syndrome by 1.06 forevery 500 ml, even though central venous/wedge pressures were not elevated; non-INHpatients were not evaluated.

Dai et al (1998)35 26, INH36, Non-INH

3.1 � 1.0ml/kg/%burn2.3 � 0.8 ml/kg/%burn

35% Review of adults with burns of �20% TBSA. Allpatients received colloid starting at 24 hours.

NA, data not available; INH, patients with inhalation injuries; Non-INH, patients without inhalation injuries.

Journal of Burn Care & Research384 Saffle May/June 2007

Modern Reports of Fluid CreepIn contrast to reports documenting increased fluidrequirements for exceptional patients, recent publica-tions have reported greatly increased requirementsfor resuscitation of a majority of routine patients with

major burn injuries. These reports are reviewed inTable 2.6–8,10,11 Some of these series document fluidvolumes far in excess of those previously reported forpatients with inhalation injuries; moreover, theyshow results from unselected series of patients,

Table 2. Review of modern reports of fluid creep

Reference

No. of PatientsWho Exceeded

ParklandRequirements

ResuscitationReceived, ml/kg/%TBSA Comments

Kaups et al (1998)6 83/83 (100%) NA Review of patients treated 1994–1995 to assessthe relationship of base deficit to outcomes.All patients exceeded Parkland calculations;the 14 patients with base deficit �6 had largerburns, more inhalation injury, highermortality, and greater fluid requirements(21 � 4 vs 12 � 3 liters, an increase of 75%).

Engrav et al (2000)11 29/50 (58%) 5.2 � 2.3 (no range given) Review from seven centers. Majority of patientexceeded Parkland requirements; this wasmore pronounced in patients withinhalation injury.

Ivy et al (2000)7 98/109 (90%) 9.36 (2.2–38.6) Prospective evaluation of the incidence ofintra-abdominal hypertension andabdominal compartment syndrome in burnpatients; seven developed the former andtwo developed the latter. Authorsrecommend routine monitoring of bladderpressure in any patient who receives �250ml/kg fluid.

Cartotto et al (2002)10 26/31 (84%) 6.7 � 2.8 Retrospective evaluation of patients treated1998–2000. Two interesting observations:first, patients arrived and began resuscitationa mean of 1.7 hours post-injury but hadalready received 2.5 � 1.9 liters of lactatedRinger’s solution. Second, Parkland formulawas quite accurate for the first 8 hourspost-burn but requirements increased afterthat in 15/31 patients.

Cancio et al (2004)59 56/89 (63%) 6.1 � 0.22 (no range given) Review of patients resuscitated 1987–1997with the modified Brooke formula, whichincluded a small amount of albumin. Burnsize and body weight were associated withincreased fluid requirements.

Friedrich et al(2004)8, Sullivanet al (2004)9

NA 3.6 � 1.1 (1970s) vs.8.0 � 2.5 (2000)

Comparison of 11 patients resuscitated during1975–1979 with 11 patients matched forage, sex, and burn size treated during 2000.Recent patients received more than doublethe fluid received by patients in the 1970sdespite equal urine output. In secondpublication, authors suggest that increasedopioid use in the first 24 hours maycontribute to increased fluid requirements.

NA, data not available.


most of whom can be presumed not to have inha-lation injury, which occurs in only 10% to 20% ofburn center admissions.

Though not systematically characterized, enoughexperience with fluid creep has been obtained to per-mit some generalization about its presentation.10 Pa-tients often arrive at burn centers having receivedsubstantial amounts of crystalloid, sometimes signif-icantly more than required, because of inaccurate es-timations of burn size or overzealous or inattentivetreatment. Parkland resuscitation is begun and con-tinues fairly smoothly until 8 to 12 hours post-burn.At that time, instead of decreasing, fluid requirementseither remain high or actually begin to escalate andrange farther and farther from predictions. As thiscontinues, problems with torso and extremity com-partment syndromes, respiratory distress, and facialswelling may develop. Requirements for large quan-tities of LR often continue unchecked despite effortsto reduce them and taper only very slowly, often re-quiring much longer than 24 hours to resolve. As anexample, Figure 1 charts the course of a 6-year-oldboy with burns of 33% TBSA who was recently resus-citated in our burn center with a Parkland-based pro-tocol. Although it could be argued that resuscitationguidelines were a bit too generous (eg, urine outputof 0.9–1.8 ml/kg/h), this does not explain the greatincrease in fluid requirements seen at 10 to 12 hourspost-burn or the lack of response in urine output untilmuch later. Up until hour 23, when fluids could fi-

nally begin to be decreased, urine output averagedonly 0.97 ml/kg/h, but the patient had already re-ceived a resuscitation volume of 5.70 ml/kg/%TBSA.

This information demonstrates that an increasingnumber of exceptions to the Parkland formula havebeen accumulating for many years. Although modernseries document a significantly greater manifestationof this trend, and in apparently unselected patientsamples, it appears likely that some mechanisms re-sponsible for fluid creep may have been influencingfluid resuscitation in burn care for much longer thanpreviously appreciated. The characteristic clinical pre-sentation described above, although not universallypresent, suggests some factors contributing to thecause of fluid creep and some potential therapeuticinterventions, as will be discussed below.

Fluid Creep and the AbdominalCompartment SyndromeIn the early 1980s, surgeons noted that increasingabdominal distension and bleeding was associatedwith oliguria and eventual renal failure, which couldbe reversed by abdominal decompression.43 Kronet al demonstrated that clinical manifestationscould be correlated with measurements of bladderpressure, which provided an objective indicationfor re-exploration.44 Since then, ACS has become awell-characterized problem. The syndrome is con-sidered secondary when it occurs in the absence of

Figure 1. Time course of fluid resuscitation for a 6-year-old boy (20 kg) with 33% TBSA scald burns. He arrived at the burn center6 hours post-injury, having received 900 ml of lactated Ringer’s solution prior to arrival. Fluid resuscitation was started according tothe Parkland formula (heavy dashed line); nurses were instructed to maintain urine output between 0.9 and 1.8 ml/kg/h (dottedline). Initial resuscitation was close to Parkland guidelines, but beginning at about 10 hours post-burn, fluid requirements increasedprogressively until about 22 hours post-burn, when urine output finally began to rise, and fluids were tapered in a stepwise manneraccording to protocol. The patient reached his calculated maintenance fluid rate of 106 ml/h at hour 36. Total resuscitation receivedwas 11.38 ml/kg/%TBSA. He had no difficulties with compartment syndromes or respiratory distress.


demonstrable intraabdominal pathology.45 Second-ary ACS in burn patients has been repeatedly de-scribed.46–50 Its occurrence has been shown to cor-relate directly with volume of crystalloid resuscitationfluid in both burn and nonburn situations.51 Theincidence is unknown, but it may be very commonin patients given sufficient amounts of fluid resus-citation.52,53 It is worthy of note that both hyper-tonic crystalloid and colloid-based resuscitation ap-pear to reduce development of ACS.54,55

Limited data suggest that other complications canalso occur from excessive resuscitation, includingmassive pleural and pericardial effusions,52 compart-mental compression in unburned extremities, and theneed to perform or prolong intubation in patientswithout inhalation injuries or facial burns.56 Re-cently, elevated intraocular pressures have also beendescribed in association with massive fluid resuscita-tion in burn patients.57,58

WHAT FACTOR(S) CAUSEFLUID CREEP?

Whereas reports of ACS and other edema-relatedcomplications demonstrate unequivocally the poten-tially disastrous consequences of fluid creep, they pro-vide little insight into its etiology. That these causesare almost certainly multiple can be inferred fromseveral observations about this phenomenon. For onething, heterogeneity in clinical practice among burncenters is the norm; despite professing adherence tothe Parkland formula, burn centers practice fluid re-suscitation differently. Both the widespread observa-tions of fluid creep and the substantial differences inabsolute fluid volumes required in different centersdocumented in Table 2 suggest that fluid creep tran-scends minor variations in resuscitation practiceand is manifested differently among these centers.A number of possible explanations can be pro-posed, which may all contribute to some extent tothe emergence of fluid creep as a clinical problem.These include the following.

The Parkland Formula Isn’t Accurate,Especially for Very Large BurnsIn Baxter’s studies, patients with large burn injuriesrequired disproportionately more fluid for resuscita-tion than those with smaller injuries.30 The majorityof patients with burns of �60% TBSA died,20 anddeath was attributed to “resuscitation failure” in amajority of cases. In a recent review of burn resusci-tation, Cancio et al found that fluid requirementscorrelated with both total and full-thickness burnsize, ranging from approximately 4.0 ml/kg/%TBSA

for moderate injuries to almost 6.0 ml/kg/%TBSAfor burns of 80% to 100% TBSA,59 and that patientswith the largest injuries were most likely to fail resus-citation attempts.60

Today mortality from burn injuries is at an all-timelow. Many patients with massive injuries survive, of-ten following aggressive resuscitation well beyondParkland confines. In turn, this success may have in-fluenced practitioners’ approach to resuscitation ofpatients with smaller injuries and encouraged lessstringent adherence to formulas. But even if this istrue, it would not explain the magnitude of increasedfluid requirements in recent reports or the “runaway”nature of fluid creep observed in many patients.

Modern Clinicians are CarelessInherent in the term fluid creep is the implication thatclinicians are permitting resuscitation to escape theircontrol through lack of attention or carelessness. Intheir review of resuscitations, Cancio et al noted thatclinicians were less likely to reduce fluid infusions inthe face of increased (�50 ml/h) urine output thanthey were to increase fluids in the face of inadequate(�30 ml/h) output.59 Nonetheless, overall urineoutputs were not excessive, which makes it unlikelythat increased fluid requirements were the result ofinattention to Parkland guidelines. Moreover, fluidcreep often continues despite directed efforts to re-duce fluid infusions. It is unquestionably true thatburn unit staff members sometimes fail to reduce flu-ids in a timely manner, but errors occur in both di-rections. Recently, a computerized protocol for fluidresuscitation proved more accurate than technician-run resuscitation in an experimental model.61 Thismay be a promising area for improving resuscitationin the future, but it appears unlikely that inadvertentover-resuscitation within burn centers explains thevast majority of fluid creep.

“Opioid Creep”In 2004, a group at Harborview Burn Center in Se-attle compared 11 patients treated during 1975 to1979 to 11 similar patients treated in 2000 and foundthat fluid requirements more than doubled for thelatter group (8.0 � 2.5 vs. 3.6 � 1.1 ml/kg/%TBSA;P � .001).8 In a follow-up publication,9 they dem-onstrated a corresponding increase in the use of nar-cotics and sedatives between these time periods:1970s patients received a mean of 3.9 � 2.2 opioidequivalents during the first 24 hours post-burn, com-pared with 26.5 � 12.3 equivalents in the patientstreated in 2000 (P � .001). Opioid dosage corre-lated with fluid requirements in these patients.These authors postulated that fluid creep is a con-


sequence of the increasing use of narcotics duringinitial burn care.

This experience exemplifies greatly increased em-phasis on the assessment and treatment of pain inhospitalized patients over the past decade in everypatient population.62 Opiates are the mainstay of paincontrol in burn patients,63,64 and these agents havesignificant cardiovascular effects. Rouby et al foundthat administration of morphine to critically ill pa-tients can “partially antagonize adrenergically medi-ated cardiovascular response to stress.”65 But evenhigh doses of narcotics appear to be well tolerated inpatients with acute burns,66 so it appears unlikely thatopiates alone could explain the dramatic magnitudeof fluid creep observed in recent years. In addition,high doses of narcotics are routinely given to patientsin a variety of other clinical conditions, without ap-parent propensity to fluid sequestration.

The Influence ofGoal-Directed ResuscitationOver the past 20 years, critical care practitioners haveattempted to adjust resuscitation to achieve the goalsof normalizing base deficit (BD) and lactic acid (LA)levels and achieving supranormal levels of cardiac in-dex (CI) and oxygen delivery (DO2) and/or con-sumption (VO2). The finding that BD and LA levelscorrelate with magnitude of injury and mortality intrauma patients67,68 and initial studies which dem-onstrated that survival correlated with attainmentof supranormal levels of CI, VO2, and DO2

69 led toprotocols to push patients to these goals, with use ofSwan-Ganz catheters, inotropes, and fluid sup-port.70,71 In initial trials this approach appeared ef-fective.72,73 However, in both meta-analyses74,75 andcarefully controlled large multicenter trials, “goal-directed” therapy has not been shown to be superiorto treatment based on standard clinical parame-ters,76,77 and it clearly requires increased volumes offluid and blood,78 with a higher incidence of ACS.79

Very similar experience has been documented inburn patients. Base deficit and LA levels correlatewith both mortality39,80 and fluid requirements forresuscitation.81 Traditional resuscitation often fails tonormalize LA and BD82 and may not reflect changesin CI and VO2.83 Goal-directed resuscitation was as-sociated with improved survival in one case-controlstudy of burn patients.84 In contrast, other groupswere unable to determine whether resuscitationaimed at normalizing BD was effective or benefi-cial,6,85 and in two other trials, attaining target valuesof preload, CI, or VO2 required more fluid—as muchas four times Parkland predictions—without obviousimprovements in survival.86,87 In a trial involving 50

adult patients randomized to receive either strictParkland or goal-directed resuscitation, Holm et alfound no differences in mortality, intensive care unitor ventilator days, pH, or serum lactate levels.88 Car-diac index was increased in the goal-directed grouponly at 24 hours post-burn, and all parameters wereidentical by 48 hours. Patients in the goal-directedgroup required 56% more fluid than Parkland pa-tients. Authors concluded that these findings “may bedue to the fact that a pure crystalloid resuscitation isincapable of restoring cardiac preload during the pe-riod of burn shock”—exactly the same conclusionreached by Baxter 25 years earlier.

It now appears clear that patients’ ability to attain atleast normal values of CI, DO2/VO2, BD, etc. ispredictive of survival following trauma and burns butthat it may be impossible to turn “nonresponders”into “responders” by specific physiologic manipula-tions.89,90 In addition, these studies confirm the clas-sic observations of Baxter and others,18 that restora-tion of preload and cardiac function and resolution ofacidosis appear to require 24 to 48 hours to occur,regardless of the resuscitation used. “Pushing” theseparameters with increased preload or inotropes re-sults in greatly increased fluid requirements withoutobvious improvements in outcome. Although use ofinvasive monitoring may still be indicated in somehigh-risk patients91 and in patients with respiratoryfailure, heart disease, or inadequate response to stan-dard treatment,75 even in these circumstances dataderived from Swan-Ganz catheters may be more use-ful in avoiding major errors than in driving resuscita-tion to specific “supraphysiologic” endpoints.

But although routine practice of goal-directed re-suscitation thus appears to be unnecessary or evenharmful, it nonetheless seems likely than many burnclinicians—who double as trauma and critical-caredoctors—have been influenced by these experiencesand may often be tempted both to measure base def-icit, lactate, and other variables and to respond toworrisome values by increasing fluid infusions,even when vital signs and urine output are ade-quate. This may contribute significantly to fluidcreep in many situations, but it still doesn’t explainthe tendency for fluid creep to persist despite at-tempts to reduce fluid infusions.

Influence of Excessive Crystalloid Infusionon Starling ForcesConsiderable evidence, both theoretical and clinical,supports the idea that excessive administration ofcrystalloid and the abandonment of colloid replenish-ment at the end of resuscitation are major contribu-tors to fluid creep.


The forces that control transcapillary fluid flux aresummarized in Starling’s equation92; their alterationsin burn injury have recently been reviewed by Dem-ling.93 The greatest edema formation occurs almostimmediately post-burn within the wound, caused bynear-total permeability to even very large (350 A)molecules, permitting leakage of fluid which is essen-tially identical to plasma.94 This effect is transient;both its duration and its magnitude are proportionalto burn size.95,96 This initial leakage of proteinslargely eliminates the oncotic pressure gradient nec-essary for maintenance of intravascular volume.Simultaneously, the densely configured collagen-hyaluronate interstitial matrix, which ordinarily actsas a “safety valve” to edema formation,97 is disrupted,increasing compliance and producing osmotically ac-tive fragments and negative (“sucking”) interstitialpressure, which also favors rapid fluid sequestra-tion.98 Although this gradient is neutralized within afew hours, compliance continues to increase as inter-stitial gel is hydrated, allowing ongoing accumulationof fluid with little change in hydrostatic pressure.99

In contrast to early edema formation, subsequentfluid sequestration occurs prominently outside thewound. Depletion of plasma proteins alone canmimic burn edema, and infusions of albumin ordextran can almost completely prevent edema inunburned tissues.100,101

With this in mind, it is easy to see that any excessivefluid given in the early post-burn period would in-crease capillary hydrostatic pressure and further re-duce oncotic pressure,102 both contributing to a cycleof accelerated capillary leakage which requires ever-greater amounts of crystalloid to satisfy. This mecha-nism could explain why fluid creep is manifested soprominently by edema in unburned tissues, includingthe abdomen, as well as the otherwise paradoxicalobservation that fluid requirements are usually fairlyclose to Parkland predictions for the first 8 hours post-injury—when capillary leakage should be greatest—but become increasingly problematic after this peri-od.10 Cancio et al found that while “resuscitationfailure” correlated with burn size, it could not bepredicted on patient characteristics alone.60 Otherfactors—including, perhaps, the volume of initialfluid therapy—appeared to play a role. This mecha-nism could explain why the fluid requirements docu-mented in some recent reports are continuing toescalate to volumes far in excess of Parkland calcula-tions, seemingly without limit.8,9 In this scenario,fluid creep becomes self-perpetuating and creates itsown physiology of edema formation.

Finally, this mechanism may also explain why fluidcreep is being reported at this point in time. Over the

past decade, three trends have emerged that all favora tendency to over-resuscitate in the early post-burnperiod. First is the practice of goal-directed resuscita-tion, discussed previously, which still influences manyclinicians. Second is the success of ubiquitous out-reach education in burn care, which has improvedburn triage but also contributed to a now-commonproblem of excessive resuscitation given by first re-sponders and inexperienced physicians, who oftengreatly overestimate burn size42,103 and sometimesrun intravenous infusions “wide open.” Thus, pa-tients often arrive at a burn center having receivedmuch of their first 8-hour Parkland requirements injust an hour or two.10

Third is the current prejudice against use of col-loid, which has developed in recent years. Severalmeta-analyses of trials comparing resuscitation reg-imens have concluded that use of colloids is dele-terious to patients in a variety of situations,104 in-cluding burn patients, for whom the odds ratio formortality with albumin usage has been calculatedto be as high as 2.40 (95% confidence limits: 1.11,5.19).105

Though widely regarded as authoritative, thesepublications have been criticized for being basedlargely on very old, heterogenous, unblinded stud-ies,106 which appears to be true for burns. Reviewsfrom the Cochran Collaboration have evaluatedonly four small trials involving burn patients,107–110

only one of which showed increased mortality withalbumin usage—the 1983 study by Goodwin etal.109 Authors found that colloid-resuscitated pa-tients required less fluid than those who receivedcrystalloid alone (2.98 vs 3.81 ml/kg/%TBSA) butalso demonstrated progressive increases in lung wa-ter up to 7 days post-burn. Mortality was higher inthe colloid group (11 of 40 patients) than in thecrystalloid group (3 of 39 patients), though all pa-tients died later of causes not obviously related tofluid resuscitation. This small study has influencedthinking about burn resuscitation for �20 yearsand has contributed, perhaps excessively, to theinterdiction of colloid use in many centers. As ref-utation of this attitude, a recent major multicentertrial of routine albumin use for resuscitation in al-most 7,000 intensive care unit patients found noincreased risk of death or other adverse out-comes.111 Burn patients were excluded from thistrial, but a more recent randomized trial involvingburn patients yielded the same conclusion andrevealed no difference in rates of multiple organfailure.112


STRATEGIES FOR PREVENTION ANDTREATMENT OF FLUID CREEP

On the basis of the information above, it can be ar-gued that an early redefinition of the Parkland for-mula that excluded colloid use, the influence of goal-directed resuscitation, and the trends that havefavored overzealous early resuscitation and the exclu-sion of colloids have combined to produce a trendtoward increased burn resuscitation requirements,which has recently been characterized as fluid creep.If so, then these mechanisms also suggest remediesthat should halt or reverse progression of this prob-lem. However, this hypothesis is unproven; the causesof fluid creep are likely multiple, so any protocol tocontrol it should address as many of these potentialcauses as possible. In addition, the magnitude of thisproblem in clinical practice, its still poorly understoodetiologies, and its serious adverse consequences arguestrongly for the performance of randomized trials toevaluate all of these therapies.

Potential therapeutic strategies include the fol-lowing:

1. Restrict Early Fluid ResuscitationAlthough the prompt institution of fluid resuscitationafter burn injury is an important contributor to im-proved survival,113 excessive initial resuscitation is alikely contributor to fluid creep, which may not beapparent until much later. As pointed out by Cancioet al, fluid requirements may fall below Parkland pre-dictions for the first few hours after injury,59 and in-fusions can often be adjusted downward during thisperiod. Close communication with first respondersand referring physicians—possibly including tele-medicine or other visual evaluation42—is essentialand helps burn professionals regulate resuscitationas soon as possible after injury. Use of widely ac-cessible programs for calculating burn size andfluid requirements may also help inexperienced cli-nicians avoid overresuscitation.114–116

2. Consider Routine Colloid, or“Colloid Rescue”Although recent literature fails to prove that colloid-based resuscitation increases mortality or complica-tions in burn patients, it also does not demonstrateany benefit of its routine use. Many patients are suc-cessfully resuscitated with crystalloid alone, withoutexcessive volumes. However, edema-related compli-cations correlate with the absolute volume of fluidsinfused; the volumes required with colloid-assistedresuscitation appear to be less than those with crys-talloids alone,32,35,59 and this has been associated

with lower abdominal pressures and incidence of clin-ical ACS.55 A few centers have also used limitedamounts of synthetic colloids such as hetastarch forroutine burn resuscitation, with good results.117,118

Thus, administration of colloid may reduce the con-sequences of fluid creep even if it does not directlyaddress its causes.

One potential approach to controlling fluid creep,therefore, would be to adhere to the original Parklandformula and infuse a colloid bolus at the end of 24hours post-burn. This is still practiced in someunits27,40 and should be considered in situationswhere resuscitation is not straightforward. Alterna-tively, some centers administer colloids to patientswho develop increasing fluid requirements during re-suscitation, as a means of “escape” from fluid creep.This mechanism likely explains the efficacy of plasma-pheresis to arrest progressive acute resuscitation fail-ure.119 Yowler and Fratianne institute resuscitationwith albumin at 12 hours post-burn when fluid re-quirements exceed 120% of normal.40 In a recentreport on burn management during Operation IraqiFreedom, Chung et al noted increasing clinical prob-lems with “resuscitation morbidity”; as a result, theyhave developed a protocol to utilize 5% albumin so-lution in any patient whose 24-fluid requirements areprojected to exceed 6 ml/kg/%TBSA.120 This hasreduced the incidence of ACS to zero.

3. Use Resuscitation ProtocolsConcerns over the occurrence of fluid creep has in-creased confusion and variation in both physician andnursing practices of fluid resuscitation in many burncenters. In response to these concerns, we have de-veloped a strict protocol for resuscitation at the Uni-versity of Utah, which includes a pathway for colloid“rescue” in patients with increasing fluid require-ments. This protocol is illustrated in Figure 2. Wehave found this helpful in reducing excessive resusci-tation and improving staff awareness of fluid resusci-tation guidelines.

4. Other Resuscitation AlternativesHypertonic saline has been used for burn resuscita-tion for decades1,121 and routinely requires less fluidthan isotonic crystalloid. In addition, Oda et al re-ported that patients resuscitated with hypertoniclactated saline had less intra-abdominal hyperten-sion than patients given LR.54 Hypertonic resuscita-tion has been recommended for use in children122

and the elderly123 who tolerated excessive fluid vol-umes poorly. However, routine hypertonic resuscita-tion carries risks of hypernatremia and hyperosmolar-ity and requires careful monitoring. Its use has been


associated with increased mortality in at least onestudy.124 A recent Cochrane Database review foundinsufficient evidence to support either beneficial ordeleterious effects of hypertonic resuscitation.125

Considerable recent research has also evaluated useof highly concentrated sodium solutions for resusci-tation. The combination of 7.5% NaCl/dextran 70(HSD) has been used for resuscitation of trauma pa-

tients in the field126 and been advocated for situationssuch as combat or mass casualties, where large vol-umes of crystalloid may be unavailable.127 In ex-perimental models, infusions of HSD producedshort-term repletion of intravascular volume whileminimizing visceral edema.128,129 However, in a clin-ical trial, a bolus of HSD did not reduce overall 24-hour fluid requirements.130 Although these data do

Figure 2. Protocol for fluid resuscitation of adult burn patients. In response to requests from nursing, this protocol wasdeveloped to permit nursing staff to manage fluid resuscitation of acute burn patients. Initial fluid rates are calculated by theParkland formula. Nurses begin hourly infusion, measure urine output, and adjust fluids according to patient response.Development of unstable vital signs, inadequate response to fluids, or persistently high fluid requirements prompt a call to thephysician. A pathway to begin colloid replacement exists for patients who display increasing fluid requirements or developevidence of torso compartment syndrome.


not clarify the value of HSD for routine burn resus-citation, they suggest that this agent might be usefulto “rescue” patients demonstrating progressive fluidcreep and edema-related complications.

Finally, it may be possible to regulate resuscitationpharmacologically. A number of investigators haveattempted to reduce the severity of burn shock byblocking specific chemical mediators of acute inflam-mation, including use of vasodilators such as hydral-azine, the serotonin antagonist ketanserin, and anti-inflammatory drugs such as hydrocortisone andibuprofen.18,93,131,132 The antioxidant vitamin C,given in high doses in the early post-burn period,has been shown to decrease fluid requirementsin clinical burn resuscitation.133 In experimentalmodels, other scavengers of oxygen radicals havealso been helpful in reducing fluid requirements forburn resuscitation.134 Although none of these ma-nipulations have found their way into widespreadclinical use, the emergence of fluid creep as a clin-ical problem may renew interest in pharmacologiccontrol of fluid resuscitation.

5. Monitor Resuscitation and ComplicationsRecent experience demonstrates that ACS is not itselfa lethal complication in burn patients, although it isoften associated with a poor prognosis from associ-ated injury.50 Early decompressive laparotomy canimprove survival markedly, especially if ACS is diag-nosed promptly.47,49 Routine monitoring of bladderpressure should be performed in patients with largeinjuries, those who demonstrate oliguria or increas-ing ventilator requirements in the face of ongoingresuscitation, or those who require excessive resusci-tation volumes, variously estimated at 6 ml/kg/%TBSA,124 250 ml/kg,7 500 ml/h,53 or 20 liters oftotal fluid.46 Established ACS is often a surgical emer-gency requiring immediate laparotomy, but impend-ing ACS, heralded by increasing bladder pressures,can be treated by escharotomy, paracentesis, or fluidrestriction. In addition, the value of colloid or hyper-tonic saline as a “rescue” therapy in such circum-stances should be evaluated.

CONCLUSIONS

The recent emergence of fluid creep as a significantclinical problem in burn treatment has promptedreview of classic investigations into the pathophys-iology of burn shock that form the foundations ofmodern burn resuscitation. In many ways, this com-plication is a result of the progress made in burn carein recent decades, as patients with larger and largerinjuries are considered salvageable and are subjected

to aggressive initial care. Pruitt has suggested “push-ing the pendulum back” in practicing fluid resuscita-tion. As an alternative, this experience should beviewed as an opportunity to move forward in burntreatment, by revisiting the principles of burn resus-citation and reevaluating current practice protocols.Although the exact causes of fluid creep remain un-determined, a number of strategies can be utilized tocontrol its magnitude and complications. This expe-rience also further underscores the need for multi-center randomized trials of resuscitation protocols todevelop the best methods of caring for severelyburned patients.

REFERENCES

1. Underhill F. The significance of anhydremia in extensive su-perficial burns. JAMA 1930;95:852–7.

2. Warden GD. Burn shock resuscitation. World J Surg 1992;16:16–23.

3. Saffle JR. The 1942 fire at Boston’s Cocoanut Grove night-club. Am J Surg 1993;166:581–91.

4. Cope O, Moore F. The redistribution of body water and thefluid therapy of the burned patient. Ann Surg 1947;126:1010–45.

5. Baxter C. Fluid volume and electrolyte changes in the earlypost-burn period. Clin Plastic Surg 1974;1:693–703.

6. Kaups KL, Davis JW, Dominic WJ. Base deficit as an indicatoror resuscitation needs in patients with burn injuries. J BurnCare Rehabil 1998;19:346–8.

7. Ivy ME, Atweh NA, Palmer J, Possenti PP, Pineau M,D’Aiuto M. Intra-abdominal hypertension and abdominalcompartment syndrome in burn patients. J Trauma 2000;49:387–91.

8. Friedrich JB, Sullivan SR, Engrav LH, et al. Is supra-Baxterresuscitation in burn patients a new phenomenon? Burns2004;30:464–6.

9. Sullivan SR, Friedrich JB, Engrav LH, et al. “Opioid creep” isreal and may be the cause of “fluid creep”. Burns 2004;30:583–90.

10. Cartotto RC, Innes M, Musgrave MA, Gomez M, CooperAB. How well does the Parkland formula estimate actual fluidresuscitation volumes? J Burn Care Rehabil 2002;23:258–65.

11. Engrav LH, Colescott PL, Kemalyan N, et al. A biopsy of theuse of the Baxter formula to resuscitate burns or do we do itlike Charlie did it? J Burn Care Rehabil 2000;21:91–5.

12. Pruitt BA. Protection from excessive resuscitation: “pushingthe pendulum back.” J Trauma 2000;49:567–8.

13. Thomas SJ, Barrow RE, Herndon D. History of the treat-ment of burns. In: Herndon D, editor. Total burn care.Philadelphia: WB Saunders; 2002. pp.1–10.

14. Evans E, Purnell O, Robinett P, Batchelor A, Martin M. Fluidand electrolyte requirements in severe burns. Ann Surg 1952;135:804.

15. Moncrief JA. Effect of various fluid regimens and pharmaco-logic agents on the circulatory hemodynamics of the imme-diate postburn period. Ann Surg 1966;164:723–52.

16. Moyer CA, Margraf HW, Monafo WW. Burn shock and ex-travascular sodium deficiency: treatment with Ringer’s solu-tion with lactate. Arch Surg 1965;90:799–811.

17. Arturson G. Pathophysiological aspects of the burn syndromewith special reference to liver injury and alterations of capillarypermeability. Acta Chir Scand Suppl 1961;Suppl 274:1–135.

18. Pruitt BA, Mason AD, Moncrief JA. Hemodynamic changesin the early postburn patient: the influence of fluid adminis-tration and of a vasodilator (hydralazine). J Trauma 1971;11:36–46.


19. Monafo WW. The treatment of burn shock by the intrave-nous and oral administration of hypertonic lactated salinesolution. J Trauma 1970;10:575–86.

20. Baxter CR, Shires T. Physiological response to crystalloidresuscitation of severe burns. Ann N Y Acad Sci 1968;150:874–94.

21. Baxter C. Fluid resuscitation, burn percentage, and physio-logic age. J Trauma 1979;19(11 Suppl):864–5.

22. Baxter CR. Problems and complications of burn shock resus-citation. Surg Clin North Am 1978;58:1313–22.

23. Shires GT, Carrico CJ, Baxter CR, Giesecke AH, Jenkins MT.Principles in treatment of severely injured patients. Adv Surg1970;4:255–324.

24. Moncrief JA. Replacement therapy. In: Artz C, Moncrief JA,Pruitt BA Jr, editors. Burns: a team approach. Philadelphia:WB Saunders; 1979. pp. 169–92.

25. Schwartz SI. Supportive therapy in burn care: consensus sum-mary on fluid resuscitation. J Trauma 1979;19(11 Suppl):876–7.

26. American Burn Association.Shock and fluid resuscitation. In:Manual of advanced burn life support. Chicago: AmericanBurn Association; 2001. pp. 33–40.

27. Warden G. Fluid resuscitation and early management. In:Herndon D, editor. Total Burn Care. Philadelphia: WBSaunders; 2002. pp. 88–97.

28. Wolf SE, Herndon D. Burns and radiation injuries. In: Mat-tox K, Feliciano D, Moore E, editors. Trauma. New York:McGraw-Hill; 2000. pp. 1137–52.

29. Pruitt BA. Fluid and electrolyte replacement in the burnedpatient. Surg Clin North Am 1978;58:1291–312.

30. Baxter CR. Guidelines for fluid resuscitation. J Trauma 1981;21:687–92.

31. Pruitt BA. Fluid resuscitation for extensively burned patients.J Trauma 1981;21:690–2.

32. Herndon DN, Barrow RE, Linares HA, et al. Inhalation in-jury in burned patients: effects and treatment. Burns InclTherm Inj 1988;14:349–56.

33. Navar PD, Saffle JR, Warden GD. Effect of inhalation injuryon fluid resuscitation requirements after thermal injury. AmJ Surg 1985;150:716–20.

34. Scheulen JJ, Munster AM. The Parkland formula in patientswith burns and inhalation injury. J Trauma 1982;22:869–71.

35. Dai NT, Chen TM, Cheng TY, et al. The comparison of earlyfluid therapy in extensive flame burns between inhalation andnoninhalation injuries. Burns 1998;24:671–5.

36. Darling GE, Keresteci MA, Ibanez D, Pugash RA, Peters WJ,Neligan PC. Pulmonary complications in inhalation injurieswith associated cutaneous burn. J Trauma 1996;40:83–9.

37. Nguyen TT, Gilpin DA, Meyer NA, Herndon DN. Currenttreatment of severely burned patients. Ann Surg 1996;223:14–25.

38. Hughes KR, Armstrong RF, Brough MD, Parkhouse N.Fluid requirements of patients with burns and inhalation in-juries in an intensive care unit. Intensive Care Med 1989;15:464–6.

39. Wolf SE, Rose JK, Desai MH, Mileski JP, Barrow RE, Hern-don DN. Mortality determinants in massive pediatric burns:an analysis of 103 children with � or � 80% TBSA burns (�or � 70% full-thickness). Ann Surg 1997;225:554–65; dis-cussion, 65–9.

40. Yowler CJ, Fratianne RB. Current status of burn resuscita-tion. Clin Plast Surg 2000;27:1–10.

41. Warner P, Connolly JP, Gibran NS, Heimbach DM, EngravLH. The methamphetamine burn patient. J Burn Care Reha-bil 2003;24:275–8.

42. Saffle JR, Edelman L, Morris SE. Regional air transport ofburn patients: a case for telemedicine? J Trauma 2004;57:57–64.

43. Richards WO, Scovill W, Shin B, Reed W. Acute renal failureassociated with increased intra-abdominal pressure. Ann Surg1983;197:183–7.

44. Kron IL, Harman PK, Nolan SP. The measurement of intra-abdominal pressure as a criterion for abdominal re-exploration. Ann Surg 1984;199:28–30.

45. Kirkpatrick AW, Balogh Z, Ball CG, Ahmed N, Chun R,McBeth P, et al. The secondary abdominal compartmentsyndrome: iatrogenic or unavoidable? J Am Coll Surg 2006;202:668–79.

46. Ivy ME, Possenti PP, Kepros J, et al. Abdominal compart-ment syndrome in patients with burns. J Burn Care Rehabil1999;20:351–3.

47. Jensen AR, Hughes WB, Grewal H. Secondary abdominalcompartment syndrome in children with burns and trauma: apotentially lethal complication. J Burn Care Res 2006;27:242–6.

48. Oda J, Ueyama M, Yamashita K, et al. Effects of escharotomyas abdominal decompression on cardiopulmonary functionand visceral perfusion in abdominal compartment syndromewith burn patients. J Trauma 2005;59:369–74.

49. Hobson KG, Young KM, Ciraulo A, Palmieri TL, Green-halgh DG. Release of abdominal compartment syndrome im-proves survival in patients with burn injury. J Trauma 2002;53:1129–33; discussion 33–4.

50. Greenhalgh DG, Warden GD. The importance of intra-abdominal pressure measurements in burned children.J Trauma 1994;36:685–90.

51. Oda J, Yamashita K, Inoue T, et al. Resuscitation fluid vol-ume and abdominal compartment syndrome in patients withmajor burns. Burns 2006;32:151–4.

52. Tekin A, Namias N, O’Keeffe T, et al. A burn mass casualtyevent due to boiler room explosion on a cruise ship: prepared-ness and outcomes. Am Surg 2005;71:210–5.

53. Britt RC, Gannon T, Collins JN, Cole FJ, Weireter LJ, BrittLD. Secondary abdominal compartment syndrome: risk fac-tors and outcomes. Am Surg 2005;71:982–5.

54. Oda J, Ueyama M, Yamashita K, et al. Hypertonic lactatedsaline resuscitation reduces the risk of abdominal compart-ment syndrome in severely burned patients. J Trauma 2006;60:64–71.

55. O’Mara MS, Slater H, Goldfarb IW, Caushaj PF. A prospec-tive, randomized evaluation of intra-abdominal pressureswith crystalloid and colloid resuscitation in burn patients.J Trauma 2005;58:1011–8.

56. Zak AL, Harrington DT, Barillo DJ, Lawlor DF, Shirani KZ,Goodwin CW. Acute respiratory failure that complicates theresuscitation of pediatric patients with scald injuries. J BurnCare Rehabil 1999;20:391–9.

57. Evans LS. Increased intraocular pressure in severely burnedpatients. Am J Ophthalmol 1991;111:56–8.

58. Sullivan SR, Ahmadi AJ, Singh CN, et al. Elevated orbitalpressure: another untoward effect of massive resuscitationafter burn injury. J Trauma 2006;60:72–6.

59. Cancio LC, Chavez S, Alvarado-Ortega M, et al. Predictingincreased fluid requirements during the resuscitation of ther-mally injured patients. J Trauma 2004;56:404–13; discussion13–4.

60. Cancio LC, Reifenberg L, Barillo DJ, et al. Standard variablesfail to identify patients who will not respond to fluid resusci-tation following thermal injury: brief report. Burns 2005;31:358–65.

61. Hoskins SL, Elgjo GI, Lu J, et al. Closed-loop resuscitation ofburn shock. J Burn Care Res 2006;27:377–85.

62. JCAHO partners to develop pain management measures. JtComm Perspect 2002;22:24.

63. MacLennan N, Heimbach DM, Cullen BF. Anesthesia formajor thermal injury. Anesthesiology 1998;89:749–70.

64. Faucher L, Furukawa K. Practice guidelines for the manage-ment of pain. J Burn Care Res 2006;27:659–68.

65. Rouby JJ, Eurin B, Glaser P, et al. Hemodynamic and meta-bolic effects of morphine in the critically ill. Circulation 1981;64:53–9.

66. Concilus R, Denson DD, Knarr D, Warden G, Raj PP. Con-


tinuous intravenous infusion of methadone for control ofburn pain. J Burn Care Rehabil 1989; 10:406–9.

67. Kaplan LJ, Kellum JA. Initial pH, base deficit, lactate, aniongap, strong ion difference, and strong ion gap predict out-come from major vascular injury. Crit Care Med 2004;32:1120–4.

68. Husain FA, Martin MJ, Mullenix PS, Steele SR, Elliott DC.Serum lactate and base deficit as predictors of mortality andmorbidity. Am J Surg 2003;185:485–91.

69. Shoemaker WC, Appel PL, Kram HB. Hemodynamic andoxygen transport responses in survivors and nonsurvivors ofhigh-risk surgery. Crit Care Med 1993;21:977–90.

70. Shoemaker WC, Appel PL, Kram HB, Bishop M, Abraham E.Hemodynamic and oxygen transport monitoring to titratetherapy in septic shock. New Horiz 1993;1:145–59.

71. Tuchschmidt JA, Mecher CE. Predictors of outcome fromcritical illness: shock and cardiopulmonary resuscitation. CritCare Clin 1994;10:179–95.

72. Fleming A, Bishop M, Shoemaker W, et al. Prospective trial ofsupranormal values as goals of resuscitation in severe trauma.Arch Surg 1992;127:1175–9; discussion 9–81.

73. Kern JW, Shoemaker WC. Meta-analysis of hemodynamicoptimization in high-risk patients. Crit Care Med 2002;30:1686–92.

74. Heyland DK, Cook DJ, King D, Kernerman P, Brun-BuissonC. Maximizing oxygen delivery in critically ill patients: amethodologic appraisal of the evidence. Crit Care Med 1996;24:517–24.

75. Kirton OC, Civetta JM. Do pulmonary artery catheters alteroutcome in trauma patients? New Horiz 1997;5:222–7.

76. Sandham JD, Hull RD, Brant RF, et al. A randomized, con-trolled trial of the use of pulmonary-artery catheters in high-risk surgical patients. N Engl J Med 2003;348:5–14.

77. Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-orientedhemodynamic therapy in critically ill patients. SvO2 Collab-orative Group. N Engl J Med 1995;333:1025–32.

78. McKinley BA, Kozar RA, Cocanour CS, et al. Normal versussupranormal oxygen delivery goals in shock resuscitation: theresponse is the same. J Trauma 2002;53:825–32.

79. Balogh Z, McKinley BA, Cocanour CS, et al. Supranormaltrauma resuscitation causes more cases of abdominal com-partment syndrome. Arch Surg 2003;138:637–42; discus-sion 42–3.

80. Kamolz LP, Andel H, Schramm W, Meissl G, Herndon DN,Frey M. Lactate: early predictor of morbidity and mortality inpatients with severe burns. Burns 2005;31:986–90.

81. Cancio LC, Galvez E, Turner CE, Kypreos NG, Parker A,Holcomb JB. Base deficit and alveolar-arterial gradient dur-ing resuscitation contribute independently but modestly tothe prediction of mortality after burn injury. J Burn Care Res2006;27:289–96; discussion 96–7.

82. Jeng JC, Lee K, Jablonski K, Jordan MH. Serum lactate andbase deficit suggest inadequate resuscitation of patients withburn injuries: application of a point-of-care laboratory instru-ment. J Burn Care Rehabil 1997;18:402–5.

83. Dries DJ, Waxman K. Adequate resuscitation of burn patientsmay not be measured by urine output and vital signs. CritCare Med 1991;19:327–9.

84. Schiller WR, Bay RC, Garren RL, Parker I, Sagraves SG.Hyperdynamic resuscitation improves survival in patientswith life-threatening burns. J Burn Care Rehabil 1997;18:10–6.

85. Choi J, Cooper A, Gomez M, Fish J, Cartotto R. The 2000Moyer Award. The relevance of base deficits after burn inju-ries. J Burn Care Rehabil 2000;21:499–505.

86. Barton RG, Saffle JR, Morris SE, Mone M, Davis B, Shelby J.Resuscitation of thermally injured patients with oxygen trans-port criteria as goals of therapy. J Burn Care Rehabil 1997;18:1–9.

87. Holm C, Tegeler J, Mayr M, Pfeiffer U, Henckel von Don-nersmarck G, Muhlbauer W. Effect of crystalloid resuscita-

tion and inhalation injury on extravascular lung water: clinicalimplications. Chest 2002;121: 1956–62.

88. Holm C, Mayr M, Tegeler J, et al. A clinical randomizedstudy on the effects of invasive monitoring on burn shockresuscitation. Burns 2004;30:798–807.

89. Velmahos GC, Demetriades D, Shoemaker WC, et al. End-points of resuscitation of critically injured patients: normal orsupranormal? A prospective randomized trial. Ann Surg2000;232:409–18.

90. Schiller WR, Bay RC, McLachlan JG, Sagraves SG. Survival inmajor burn injuries is predicted by early response to Swan-Ganz-guided resuscitation. Am J Surg 1995;170:696–9; dis-cussion 9–700.

91. Friese RS, Shafi S, Gentilello LM. Pulmonary artery catheteruse is associated with reduced mortality in severely injuredpatients: a National Trauma Data Bank analysis of 53,312patients. Crit Care Med 2006;34:1597–601.

92. Starling E. On the absorption of fluids from the connectivetissue spaces. J Physiol (London) 1986;19:312–26.

93. Demling RH. The burn edema process: current concepts.J Burn Care Rehabil 2005;26:207–27.

94. Arturson G. Microvascular permeability to macromoleculesin thermal injury. Acta Physiol Scand Suppl 1979;463:111–22.

95. Carvajal HF, Linares HA, Brouhard BH. Relationship ofburn size to vascular permeability changes in rats. Surg Gy-necol Obstet 1979;149:193–202.

96. Birke G, Liljedahl SO, Plantin LO. Distribution and losses ofplasma proteins during the early stage of severe burns. Ann NY Acad Sci 1968;150:895–904.

97. Guyton AC. Interstitial fluid pressure. II. Pressure-volumecurves of interstitial space. Circ Res 1965;16:452–60.

98. Leape LL. Initial changes in burns: tissue changes in burnedand unburned skin of rhesus monkeys. J Trauma 1970;10:488–92.

99. Miserocchi G, Negrini D, Passi A, De Luca G. Developmentof lung edema: interstitial fluid dynamics and molecular struc-ture. News Physiol Sci 2001;16:66–71.

100. Demling RH, Kramer GC, Gunther R, Nerlich M. Effect ofnonprotein colloid on postburn edema formation in soft tis-sues and lung. Surgery 1984;95:593–602.

101. Demling RH, Kramer G, Harms B. Role of thermal injury-induced hypoproteinemia on fluid flux and protein perme-ability in burned and nonburned tissue. Surgery 1984;95:136–44.

102. Zetterstrom H, Arturson G. Plasma oncotic pressure andplasma protein concentration in patients following thermalinjury. Acta Anaesthesiol Scand 1980;24:288–94.

103. Hammond JS, Ward CG. Transfers from emergency room toburn center: errors in burn size estimate. J Trauma 1987;27:1161–5.

104. Schierhout G, Roberts I. Fluid resuscitation with colloid orcrystalloid solutions in critically ill patients: a systematic re-view of randomised trials. BMJ 1998:28;316:961–4.

105. Alderson P, Bunn F, Lefebvre C, et al. Human albumin so-lution for resuscitation and volume expansion in critically illpatients. Cochrane Database Syst Rev 2002:CD001208.Available at: http://www.cochrane.org/reviews/en/ab001208.html; Internet; accessed January, 2007.

106. Webb AR. The appropriate role of colloids in managing fluidimbalance: a critical review of recent meta-analytic findings.Crit Care 2000;4(Suppl 2):S26–32.

107. Jelenko C Wheeler ML, Callaway BD, Divilio LT, BucklenKR, Holdredge TD. Shock and resuscitation. II: Volumerepletion with minimal edema using the “HALFD” (Hyper-tonic Albuminated Fluid Demand) regimen. JACEP 1978;7:326–33.

108. Greenhalgh DG, Housinger TA, Kagan RJ, et al. Mainte-nance of serum albumin levels in pediatric burn patients: aprospective, randomized trial. J Trauma 1995;39:67–73.

109. Goodwin CW, Dorethy J, Lam V, Pruitt BA. Randomized


trial of efficacy of crystalloid and colloid resuscitation on he-modynamic response and lung water following thermal in-jury. Ann Surg 1983;197:520–31.

110. Recinos PR, Hartford CA, Ziffren SE. Fluid resuscitation ofburn patients comparing a crystalloid with a colloid contain-ing solution: a prospective study. J Iowa Med Soc 1975;65:426–32.

111. Finfer S, Bellomo R, Boyce N, French J, Myburgh J, NortonR. A comparison of albumin and saline for fluid resuscitationin the intensive care unit. N Engl J Med 2004;350:2247–56.

112. Cooper AB, Cohn SM, Zhang HS, Hanna K, Stewart TE,Slutsky AS. Five percent albumin for adult burn shockresuscitation: lack of effect on daily multiple organ dysfunc-tion score. Transfusion 2006;46:80–9.

113. Barrow RE, Jeschke MG, Herndon DN. Early fluid resusci-tation improves outcomes in severely burned children. Resus-citation 2000;45:91–6.

114. Roth AC, Leon MA, Milner SM, Herting RL, Hahn AW. Apersonal digital assistant for determination of fluid needs forburn patients. Biomed Sci Instrum 1997;34:186–90.

115. Milner SM, Rylah LT, Bennett JD. The burn wheel: a prac-tical guide to fluid resuscitation. Burns 1995;21:288–90.

116. Neuwalder JM, Sampson C, Breuing KH, Orgill DP. A re-view of computer-aided body surface area determination:SAGE II and EPRI’s 3D Burn Vision. J Burn Care Rehabil2002;23:55–9; discussion, 4.

117. Waters LM, Christensen MA, Sato RM. Hetastarch: an alter-native colloid in burn shock management. J Burn Care Re-habil 1989;10:11–6.

118. Waxman K, Holness R, Tominaga G, Chela P, Grimes J.Hemodynamic and oxygen transport effects of pentastarch inburn resuscitation. Ann Surg 1989;209:341–5.

119. Warden GD, Stratta RJ, Saffle JR, Kravitz M, Ninnemann JL.Plasma exchange therapy in patients failing to resuscitatefrom burn shock. J Trauma 1983;23:945–51.

120. Chung KK, Blackbourne LH, Wolf SE, et al. Evolution ofburn resuscitation in operation Iraqi freedom. J Burn CareRes 2006;27:606–11.

121. Monafo WW, Halverson JD, Schechtman K. The role of con-centrated sodium solutions in the resuscitation of patientswith severe burns. Surgery 1984;95:129–35.

122. Bowser-Wallace BH, Caldwell FT A prospective analysis ofhypertonic lactated saline v. Ringer’s lactate-colloid for theresuscitation of severely burned children. Burns Incl ThermInj 1986;12:402–9.

123. Bowser-Wallace BH, Cone JB, Caldwell FT. Hypertonic lac-

tated saline resuscitation of severely burned patients over 60years of age. J Trauma 1985;25:22–6.

124. Huang PP, Stucky FS, Dimick AR, Treat RC, Bessey PQ, RueLW. Hypertonic sodium resuscitation is associated with renalfailure and death. Ann Surg 1995;221:543–54; discussion54–7.

125. Bunn F, Roberts I, Tasker R, Akpa E. Hypertonic versus nearisotonic crystalloid for fluid resuscitation in critically ill pa-tients. Cochrane Database Syst Rev 2004:CD002045. Avail-able at: http://www.cochrane.org/reviews/en/subtopics/74.html; Internet; accessed January, 2007.

126. Vassar MJ, Perry CA, Gannaway WL, Holcroft JW. 7.5%sodium chloride/dextran for resuscitation of trauma patientsundergoing helicopter transport. Arch Surg 1991;126:1065–72.

127. Thomas SJ, Kramer GC, Herndon DN. Burns: military op-tions and tactical solutions. J Trauma 2003;54(5 Suppl):S207–18.

128. Kinsky MP, Milner SM, Button B, Dubick MA, Kramer GC.Resuscitation of severe thermal injury with hypertonic salinedextran: effects on peripheral and visceral edema in sheep.J Trauma 2000;49:844–53.

129. Elgjo GI, Poli de Figueiredo LF, Schenarts PJ, Traber DL,Traber LD, Kramer GC. Hypertonic saline dextran producesearly (8–12 hrs) fluid sparing in burn resuscitation: a 24-hrprospective, double-blind study in sheep. Crit Care Med2000;28:163–71.

130. Murphy JT, Horton JW, Purdue GF, Hunt JL. Cardiovascu-lar effect of 7.5% sodium chloride-dextran infusion after ther-mal injury. Arch Surg 1999;134:1091–7.

131. Bjork J, Arturson G. Effect of cimetidine, hydrocortisonesuperoxide dismutase and catalase on the development ofoedema after thermal injury. Burns Incl Therm Inj 1983;9:249–56.

132. Holliman CJ, Meuleman TR, Larsen KR, et al. The effect ofketanserin, a specific serotonin antagonist, on burn shockhemodynamic parameters in a porcine burn model. J Trauma1983;23:867–71.

133. Tanaka H, Matsuda T, Miyagantani Y, Yukioka T, MatsudaH, Shimazaki S. Reduction of resuscitation fluid volumes inseverely burned patients using ascorbic acid administration: arandomized, prospective study. Arch Surg 2000;135:326–31.

134. Horton JW, Burton KP, White DJ. The role of toxic oxygenmetabolites in a young model of thermal injury. J Trauma1995;39:563–9.


The Phenomenon of “Fluid Creep” in Acute Burn … Phenomenon of “Fluid Creep” in Acute Burn Resuscitation ... The development of effective fluid resuscitation reg- ... Journal

Documents